Preface & Task Force Recommendations Welcome The End of Cheap Oil Laney's Review of Loudat's 1997 Report Loudat's 2000 Report on Economic Impacts of Hawaii's Energy Tax Credit California's Renewable Energy Program Renewable Energy Policies in Other States North Carolina's Energy Programs Arizona Energy Policies Federal Policies Arithmetic, Population, and Energy Honolulu Community Action Program Solar Water Systems in Self-Help Housing in Waianae Solar in Hawaii HECO's Energy $olutions Program Priming the Energy Pump in Hawaii Being Cool at Iolani School

Publications > Proceedings of the Energy Symposium A Peer Review of The Economic and Fiscal Impacts of the Hawaii Energy Conservation Income Tax Credit By Thomas A. Loudat, Ph.D., Revised January 27, 1997

Dr. Leroy Laney Professor of Economics and Finance, Hawaii Pacific University Dr. Laney joined the HPU faculty as a Full Professor in 1998, after a broad ranging background in banking, government, and other academic positions. He has served as a Staff Economist on the President’s Council of Economic Advisers in Washington, D.C., as an International Economist with the U.S. Treasury in Washington, and as a Senior Economist in the Federal Reserve System. His academic experience includes adjunct teaching positions at the University of Colorado at Boulder, Southern Methodist University, the University of Texas (Arlington and Dallas campuses), and the University of Hawaii at Manoa. He also held a joint appointment in the Economics and Finance Departments at Butler University in Indianapolis during the 1989-1990 academic year, and served there as Chairman of the Economics Department. From 1990 through 1998, his position was Senior Vice President and Chief Economist for First Hawaiian Bank in Honolulu. There he held a highly visible position as the Bank’s spokesman on all international, national, and local economic matters. He has served as Chairman of the Council on Revenues of the State of Hawaii, and as a member of the Market Research Committee of the Hawaii Visitors and Convention Bureau. Professor Laney has published widely in academic journals, Federal Reserve and other bank publications, and in edited volumes and conference proceedings. He remains a Consultant to First Hawaiian Bank, and has also consulted with the University of Hawaii Economic Research Organization, Alexander & Baldwin, Inc., Matson Navigation Company, Colliers Monroe and Friedlander, The Estate of James Campbell, the Waikiki Improvement Association, the State of Hawaii, and the County of Hawaii. He is a member of Who’s Who in America, Who’s Who in the World, Who’s Who in Finance and Industry, Who’s Who in the West, the American Economic Association, the Western Economic Association, the National Association for Business Economics, and Lambda Alpha real estate honorary. His Ph.D. in Economics is from the University of Colorado at Boulder. He also holds a Bachelor of Industrial Engineering from Georgia Tech, and an MBA in accounting and finance from Emory University. From 1967 to 1971, he served as a Lieutenant in the U.S. Navy Supply Corps. Dr. Laney is married to the former Sandra Prescott of Atlanta, and has two sons.

Introduction

This peer review of the subject paper has been commissioned by the Strategic Industries Division of the State of Hawaii’s Department of Business, Economic Development, and Tourism (DBED&T). The overall purpose is to provide an objective critique of Dr. Thomas Loudat’s earlier paper for an Energy Symposium held at Hawaii Electric Company on November 9, 2000. This review paper will proceed to discuss the Loudat paper in the order that it is written. Then some supplementary empirical results from the author’s own investigations will be presented, and finally some thoughts on Hawaii’s energy policy options will be offered.

Overview of Findings

The purpose of the Loudat paper is to provide a quantitative assessment of the impact of the State of Hawaii’s Energy Conservation Income Tax Credit (ECITC) on investment in solar energy systems. This tax credit has been in effect since 1977, even though the percentage of the tax credit allowed has varied over time. Upon its introduction in 1977, the credit was 10 percent. Then it was raised to 50 percent in the years 1978 through 1985. As oil prices collapsed in the mid-1980s, the credit was lowered to 10 percent again for one year in 1986. Since then, the credit was15 percent over the 1987-1989 period, and it has been kept at 35 percent since 1990. It is fairly obvious that the amount of the credit has been influenced by the level of overall energy prices throughout its existence. Loudat begins his paper by emphasizing the investment nature of the decision to purchase a solar system, projecting benefits out over a 25-year life span of a given system. Then, in an Executive Summary, he lists research assumptions and major conclusions. These conclusions include findings of positive fiscal and employment impacts of the ECITC program, plus an estimate that the solar industry will shrink to 59 percent of its current size if the ECITC is eliminated, even if Hawaiian Electric’s current Demand Side Management (DSM) program remains in place. In contrast, if both the ECITC and DSM remain, the industry is projected to grow by up to 70 percent. He further estimates that the State government’s ECITC expenditures generate ten times that amount of overall economic output, one job per ten installed systems, and labor income of over three times the original ECITC expenditure. These estimated benefits naturally depend on the assumed level of future energy prices. An oil crisis such as occurred during the 1975-1985 period is calculated to increase the above estimated economic and fiscal impacts by 20 to 300 percent, depending on when such crisis occurs during the life of a system. Loudat’s analysis does not attempt to measure the avoided negative externalities of continuing to burn fossil fuels. He does mention that such externalities are particularly important to an economy like Hawaii, where tourism and the environment are of such critical importance.

Summary of Analysis

It is not possible to recount all of the detailed analysis of the Loudat paper itself here; the reader is referred to that paper for those details. This section briefly reviews the highlights and assumptions of that analysis, commenting upon them where that is appropriate. Basically, Loudat uses the State of Hawaii Input/Output model published and maintained by DBED&T to assess the economic and fiscal impacts of both costs and benefits of the ECITC expenditure. Purchase of the solar system is viewed as a 25-year investment, and Loudat considers alternative impacts if the system is cash-financed versus borrowing-financed. (If a system is borrowing-financed, overall economic benefits improve slightly but are shifted to later years.) The economic benefit of a solar system is the stimulus it provides to an individual to purchase a solar system, as well as this purchase’s consequent economic and fiscal impacts. The costs of the ECITC are the economic and fiscal impacts of purchasing fossil fuel generated energy, foregone due to the purchase of a solar system. If the ECITC is eliminated, other economic and fiscal costs would be incurred due to the estimated reduction in the size of the solar industry. Total economic and fiscal impacts of the ECITC are calculated by multiplying the per system impacts by the estimated number of systems. This estimated number of systems depends not only on the size of the ECITC, but also on the supplemental help of the DSM program. An oil crisis, such as occurred between 1975 and 1985, would cause electricity rate increases much greater than assumed in the base case scenario. Such rate increases mean additional energy costs savings to purchasers of solar systems, as well as added positive economic and fiscal impacts. If the oil crisis occurs early in the life of the system purchased, these positive impacts will be greater than if the crisis occurs later in its life. The reader is referred to the paper for specific assumptions and conclusions from them. Loudat concludes his description of the analysis by outlining in detail several economic and fiscal impacts not measured by the analysis. Understandably, most of these impacts would be difficult to quantify: There would be a negative impact on Hawaii’s position as a Pacific Basin energy development and implementation leader. (Hawaii has the highest per capita number of solar systems in the nation.) There would likely be a negative impact on business investment in Hawaii due to vacillating state policy, which reduces certainty of return on that investment. Negative impacts of such things as unemployment insurance costs and retraining are not included. Positive impacts of permit fees and property tax revenues are not measured. Positive externalities from reduced oil consumption are not included. (If the cost of these negative consequences were incorporated into the price of oil, the energy costs savings would be significantly larger. And the larger the energy costs savings, the larger are the positive economic and fiscal impact of the ECITC.)

Critique of the Paper

As any economist who has ever conducted an analysis such as that presented in the Loudat study knows, conclusions are often very sensitive to the assumptions made. Yet, the analyst is forced to make many such assumptions in order to proceed with the analysis. This particular paper might be called into question because the study was conducted for the Hawaii Solar Energy Association. Still, this reviewer finds the assumptions and conclusions from them to be reasonable and sound. Furthermore, the analysis appears to have been conducted carefully and in great detail. This does not deny that other analyses, with other assumptions, might reach different conclusions. Yet in the absence of other work, the burden of proof is still upon those who challenge the results of the current paper. Loudat is currently preparing an updated and revised version of the paper reviewed here. That revised paper may include other salient points that either reinforce or diminish the findings of this reviewed paper. Upon this writing, this reviewer has not seen the revised paper.

Regression Analysis of the Impact of Solar Tax Credits

While the above assumptions and conclusions are important in assessing the total net economic and fiscal impacts of the ECITC, the critical question is: Are solar tax credits effective in stimulating investment in solar energy systems? An answer to this question is important because individuals might be motivated, at least to a certain extent, by other external circumstances to invest in solar systems even without the ECITC. For example, just the existence of higher energy prices alone could motivate a decision to invest in a solar system, because savings would exist even without a tax credit such as the ECITC. And clearly the percentage amount of the tax credit has varied with the level of energy prices over the life of the credit, so effects might potentially be hard to attribute to individual causal factors. Any public policy decision has to consider the incremental impact of that policy over and above what would occur just because of existing market forces. Loudat attempts to address the causal impact of the tax credit on solar systems sold with a regression analysis presented on paper 18 of his paper. In that regression, one independent variable – the percentage amount of the solar tax credit – is regressed upon a dependent variable specified as the annual number of systems sold. The interval of available data at the time of the paper was 1977 to 1992. The outcome of the regression suggests a high degree of causal impact. The adjusted R-squared is .73, and the t-statistic on the independent variable, at 6.37, is highly significant. By regressing his data in double log formulation, Loudat is able to interpret the coefficient on the single independent variable as an elasticity. The value of that coefficient indicates that, on average over the life of the ECITC, every one percent increase in the amount of the tax credit results in a 1.5 percent increase in the number of systems sold. While this outcome constitutes tangible evidence that the credit has indeed been effective in stimulating investment in solar systems, it might be considered by some to be incomplete. That is, did investors perhaps purchase solar systems just because energy prices were higher, not so much because of the existence of the tax credit? In addition, if the purchase was borrowing financed, did the level of interest rates affect the purchase decision? These questions cannot be addressed directly with the existing regression work in the paper. This reviewer undertook further regression analysis to address such questions specifically. Results are presented in Exhibits I through III. Exhibit I is simply a replication of Loudat’s regression, with the tax credit as the single independent variable and systems sold as the dependent variable. Outcomes are indeed the same as in the Loudat paper.

Exhibit I.
year	systems sold		tax credit
1977	1101	7.003974137	10	2.302585093
1978	4061	8.309184528	50	3.912023005
1979	4375	8.383661799	50	3.912023005
1980	4704	8.45616849	50	3.912023005
1981	6445	8.771059915	50	3.912023005
1982	4407	8.390949465	50	3.912023005
1983	3148	8.05452261	50	3.912023005
1984	4464	8.403800504	50	3.912023005
1985	6740	8.815815204	50	3.912023005
1986	592	6.383506635	10	2.302585093
1987	354	5.869296913	15	2.708050201
1988	316	5.755742214	15	2.708050201
1989	327	5.789960171	15	2.708050201
1990	1180	7.073269717	35	3.555348061
1991	1314	7.180831199	35	3.555348061
1992	1261	7.139660336	35	3.555348061

SUMMARY OUTPUT

Regression Statistics
Multiple R		0.862287047
R Square		0.743538951
Adjusted R Square		0.725220304
Standard Error		0.576523042
Observations		16

ANOVA
	df	SS	MS	F	Significance F
Regression	1	13.49098579	13.49098579	40.58918631	1.74E-05
Residual	14	4.653303458	0.332378818
Total	15	18.14428924

	Co efficients	Standard Error	t Stat	P-value	Lower 95%	Upper 95%	Lower 95.0%	Upper 95.0%
Intercept	2.364748	0.816713	2.895443	0.011747	0.613069	4.116427	0.613069	4.116427
tax credit	1.498319	0.235179	6.370964	1.74E-05	0.993909	2.002729	0.993909	2.002729

Exhibit II adds two other variables to Loudat’s regression – the average annual price per barrel of crude oil and an interest rate. In this exhibit, the interval was kept the same for comparison to Loudat’s results. Again, data regressed was in log-linear form.